What it does
An n8n workflow that watches a Notion database for new pages and reads their text blocks, leaving out images and videos. The text is joined, split into chunks and tagged with metadata such as page ID, creation time and title. Google Gemini creates embeddings for each chunk, and those embeddings are stored in a Pinecone vector store for search.
Use cases
- 01Let an AI assistant search new Notion pages
- 02Index an internal wiki for document retrieval
- 03Keep a Pinecone index updated as pages are added
Connects
Notion · Token Splitter · Pinecone Vector Store · Default Data Loader · Embeddings Google Gemini